Rheumatoid arthritis (RA) is a chronic inflammatory erosive polyarthritis of unknown etiology. Predisposition to RA is inherited as a complex trait. Although specific HLA-DRB1 alleles show statistically significant increased frequencies among patients with RA that vary with race and ethnicity, this genetic association does not adequately explain susceptibility to RA in African-Americans. We hypothesize that other genes modulate RA risk in African-Americans. The HLA-DRB1 association with RA risk is consistent with the role of adaptive immunity in predisposition to RA. Molecular mechanisms that govern innate immune responses are also hypothesized to participate in the pathogenesis of RA . MICA and Fc gammaRIllA (FcgammaRIIIA; CD16) allelic associations with RA susceptibility provide evidence for the involvement of two different pathways of innate immune function. Molecular interactions in other activation and signaling pathways of innate immunity, involving LBP, CD14, MD2, MYD88, IKBL and pattern recognition receptors (PRR) for microbial pathogens, such as Toll-like receptor 4 (TLR-4), as well as those affected by the TH1 enhancer interleukin 18 (IL-18), may also modulate RA risk. There are gaps in our knowledge about MICA and FcgammaRIIIA polymorphisms, known determinants of RA susceptibility, or whether molecules involved in microbial signaling pathways or IL-18 modulate the risk of RA in African-Americans. We hypothesize: HLA-DRB1 alleles encoding the shared epitope interact with genetic determinants of IL-18 and molecules involved in pathways of innate immune function (MICA, Fc(RIIIA, LBP, CD14, TLR4, MD2, MYD88, IKBL) to modulate RA risk in African-Americans. DNA from well-characterized samples of African-Americans with RA in Boston and Atlanta, healthy African-American controls from Boston and the African-American national repository will be used in high-throughput molecular techniques for the determination of microsatellites and to construct single nucleotide polymorphism (SNP) haplotypes and multi-locus genotypes to test this hypothesis. Our specific aim is to determine the multi-locus genotype that best predicts predisposition to RA in African-Americans.